`__________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________________
`
`MYLAN PHARMACEUTICALS INC.,
`Petitioner,
`
`v.
`
`MERCK SHARP & DOHME CORP.,
`Patent Owner.
`__________________
`
`Case IPR2020-00040
`U.S. Patent 7,326,708
`__________________
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`
`
`
`
`
`Merck Exhibit 2140, Page 1
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`I, Rebecca Leigh Shultz, Ph.D., hereby declare as follows:
`
`I.
`
`INTRODUCTION & BACKGROUND
`
`1.
`
`I am an employee of Merck Sharp & Dohme Corp. (“Merck”) and
`
`currently serve as Merck’s Associate Vice President, Global Project & Alliance
`
`Management. I have been a Merck employee since 2001.
`
`2.
`
`I joined Merck shortly after receiving my Ph.D. in inorganic
`
`chemistry from the University of North Carolina at Chapel Hill. I received my
`
`B.S. in chemistry from the University of Florida in 1995. From 2001 to 2004, I
`
`was a Senior Research Chemist in the Pharmaceutical Research & Development
`
`Department (“PR&D”) of Merck Research Laboratories (“MRL”).
`
`3.
`
`In late 2001, I joined the project team at Merck responsible for
`
`developing an inhibitor of dipeptidyl peptidase-IV (“DPP-IV”) into a treatment for
`
`type 2 diabetes. Over the course of the DPP-IV project, I led a functional sub-team
`
`responsible for the physicochemical characterization of candidate compounds that
`
`had been identified through Merck’s drug discovery program as well as evaluating
`
`their performance in proposed formulations for clinical studies. In this role, I
`
`personally performed many analytical tests and formulation studies and also
`
`reviewed the results of experiments performed by other scientists on the DPP-IV
`
`project. As such, I have first-hand knowledge of the data generated by the DPP-IV
`
`project team over the course of sitagliptin’s development.
`
`2
`
`Merck Exhibit 2140, Page 2
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`4.
`
`One of the lead compounds at the time I joined the DPP-IV project
`
`was sitagliptin, which had received the internal Merck designation “L-224715.”
`
`Subsequently, during Phase III development, the compound also received the
`
`designation “MK-0431.” Merck’s research and development of sitagliptin, and the
`
`work of the DPP-IV project team, culminated in an FDA-approved dosage form of
`
`sitagliptin for the treatment of type 2 diabetes, which Merck markets today under
`
`the tradename JANUVIA®.
`
`5.
`
`Sitagliptin’s formal chemical name is 4-oxo-4-[3-(trifluoromethyl)-
`
`5,6-dihydro[1,2,4] triazolo[4,3-a]pyrazine-7(8H)-yl]-1-(2,4,5-trifluorophenyl)
`
`butan-2-amine and the structural formula of the compound is shown below:
`
`Figure 1. Chemical structure of sitagliptin (L-224715 or MK-0431).
`
`6.
`
`The active pharmaceutical ingredient (“API”) in JANUVIA® is a
`
`
`
`crystalline monohydrate of a phosphoric acid salt of sitagliptin in which the
`
`dihydrogenphosphate (“DHP”) counterion from phosphoric acid and sitagliptin are
`
`present in a 1:1 stoichiometric ratio.
`
`
`
`3
`
`Merck Exhibit 2140, Page 3
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`7.
`
`I understand that Merck is the owner and assignee of U.S. Patent No.
`
`7,326,708 (“the ’708 patent”) and that the subject matter of the ’708 patent is
`
`generally directed to the 1:1 DHP salt of sitagliptin and its crystalline monohydrate
`
`form. I further understand that the named inventors of the ’708 patent are Stephen
`
`Howard Cypes, Alex Minhua Chen, Russell R. Ferlita, Karl Hansen, Ivan Lee,
`
`Vicky K. Vydra, and Robert M. Wenslow. The named inventors and I were all
`
`members of the DPP-IV project team and worked closely together to develop
`
`sitagliptin, including with respect to selecting the specific salt and crystal form of
`
`sitagliptin used in JANUVIA®.
`
`8.
`
`In this declaration, I provide facts based on my personal knowledge
`
`regarding Merck’s research and development of sitagliptin, including the
`
`contributions of the named inventors of the ’708 patent to the DPP-IV project, the
`
`synthesis and characterization of sitagliptin’s salt and crystal forms, the discovery
`
`of the crystalline monohydrate of the 1:1 DHP salt, and its selection as the solid
`
`form of sitagliptin used in the final market formulation for JANUVIA®. I have
`
`first-hand knowledge concerning the work of the DPP-IV project team and am
`
`familiar with the information that became known to the team and how key
`
`decisions in the project were made.
`
`9. Members of the DPP-IV project team, including the inventors and
`
`myself, recorded experimental observations in laboratory notebooks issued by
`
`4
`
`Merck Exhibit 2140, Page 4
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`Merck. I am familiar with Merck’s practices and procedures with respect to
`
`laboratory notebooks. In the usual and ordinary course of its business, and at the
`
`time of the DPP-IV project, Merck issued numbered laboratory notebooks to
`
`scientists for the purpose of recording their daily research activity. Each laboratory
`
`notebook page was individually numbered, and shorthand references to lab
`
`notebooks include both the lab notebook number as well as the page number, for
`
`example, “NB 60659-110.” Sample materials and experimental procedures often
`
`incorporate such shorthand references so that other project team members can
`
`easily identify the source of the sample or procedure in question. In accordance
`
`with Merck’s standard practices, I recorded the entries in my lab notebook at or
`
`near the time that I ran my experiments.
`
`10.
`
`I maintained several lab notebooks for my work on the DPP-IV
`
`project, true and correct excerpts of which may be found in the following exhibits:
`
`
`
`
`
`
`
`
`
`
`
`EX2141 – Lab Notebook (“LNB”) 60659
`
`EX2142 – Supplemental Data for LNB 60659
`
`EX2143 – LNB 26180
`
`EX2144 - Supplemental Data for LNB 26180
`
`EX2145 – LNB 72917
`
`
`
`5
`
`Merck Exhibit 2140, Page 5
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`11.
`
`In the usual and ordinary course of Merck’s business, project teams
`
`regularly produce certain milestone documents for the purpose of documenting
`
`data and related conclusions that enable key decisions as part the usual and
`
`ordinary course of research and development at Merck. The data compiled in these
`
`reports are collected and communicated by the Merck scientists responsible for
`
`(and with knowledge of) the analytical tests and experiments from which the data
`
`were generated. I was personally involved in drafting several of these documents
`
`as a member of the DPP-IV project team, true and correct copies of which may be
`
`found in the following exhibits:
`
`
`
`
`
`
`
`
`
`
`
`EX2146 – Preliminary Pharmaceutical Assessment of L-224715
`(“Dec. 21, 2001 Preliminary Assessment”)
`
`EX2147 – Pharmaceutical Evaluation of L-224715 (“Apr. 9, 2002
`Pharmaceutical Evaluation”)
`
`EX2148 – L-000224715 Preformulation Report (“Sept. 30, 2002
`Preformulation Report”)
`
`EX2149 – Physico-chemical characteristics of L-000224715
`phosphate salt, monohydrate form (“June 30, 2004 Monohydrate
`Report”).
`
`EX2123 – L-000224715 (MK-0431) Preliminary Market Formulation
`Development Report (“Aug. 31, 2005 Formulation Development
`Report”)
`
`12. Finally, MRL maintains a database of periodic progress reports
`
`generated in the usual and ordinary course of Merck’s business to track the work of
`
`project teams and scientists. The data in these reports were recorded at or near the
`
`6
`
`Merck Exhibit 2140, Page 6
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`time they were generated by the responsible Merck scientist, or communicated by
`
`the Merck scientist responsible for (and with knowledge of) the analytical tests or
`
`experiments from which the data were generated. True and correct copies of MRL
`
`progress reports submitted by me for December 2001 and January, February,
`
`March, and September of 2002 may be found in EX2150, EX2151 EX2152,
`
`EX2153, and EX2154
`
`II.
`
`SYNTHESIS AND CHARACTERIZATION OF SITAGLIPTIN SALTS
`
` Characterization of the Sitagliptin Freebase
`
`13. To evaluate the suitability of the sitagliptin freebase for development,
`
`I conducted experiments in December 2001 to investigate the physical properties
`
`of the freebase including its particle morphology, thermal properties, and
`
`hygroscopicity. I also examined the freebase’s chemical stability in solution.
`
`1.
`
`Particle Morphology
`
`Figure 2. Microscope image of sitagliptin freebase.
`
`
`
`7
`
`Merck Exhibit 2140, Page 7
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`14.
`
`I examined the particle morphology of the sitagliptin freebase using a
`
`microscope. As shown in Figure 2, the particles of the freebase were needle-like,
`
`with an aspect ratio of about 10:1 and length between about 50 and 100 µm. See
`
`EX2141 (LNB 60659-109) at 15; see also EX2146 (Dec. 21, 2001 Preliminary
`
`Assessment) at 1, 3. Avoiding particles with needle-like morphology (as in the
`
`case of the freebase) was an important early goal for the DPP-IV project, as
`
`powders formed from particles with needle or needle-like morphology (or other
`
`high-aspect-ratio particle morphologies) generally have poor flow characteristics,
`
`which limits their use in dry formulations.
`
`2.
`
`TGA/DSC
`
`15.
`
`I investigated the thermal properties of the sitagliptin freebase using
`
`thermogravimetric analysis (“TGA”) and differential scanning calorimetry
`
`(“DSC”) using a 10°C/min heating rate. The TGA and DSC traces I obtained are
`
`shown in Figure 3 and Figure 4 below, and indicate the presence of a single
`
`crystalline polymorph with an onset of melting at 116.80°C and a peak melting
`
`temperature of 118.01°C. See EX2141 (LNB 60659-109) at 15; EX2146 (Dec. 21,
`
`2001 Preliminary Assessment) at 3.
`
`8
`
`Merck Exhibit 2140, Page 8
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`
`
`
`
`Figure 3. TGA trace of the sitagliptin freebase.
`
`
`
`Figure 4. DSC trace of the sitagliptin freebase.
`
`9
`
`Sample: L224715_fb_tga
`Size: 3.6000 mg
`Method: standard method
`110
`
`TGA
`
`File: C:...\TGA\L224715_fb_tga
`Operator: LS
`Run Date: 4-Dec-01 14:45
`
`0.3858% Loss before 150 deg C
`(0.01389mg)
`
`50
`
`100
`
`150
`Temperature (°C)
`
`200
`
`250
`
`300
`
`Universal V2.3C TA Instruments
`
`100
`
`90
`
`80
`
`70
`
`Weight (%)
`
`60
`
`0
`
`Sample: L224715_freebase_dsc
`Size: 1.3200 mg
`Method: standard
`1
`
`116.80°C
`102.7J/g
`
`DSC
`
`File: C:...\L224715_fb_DSC.000
`Operator: LS
`Run Date: 4-Dec-01 15:31
`
`223.37°C
`107.0J/g
`
`255.39°C
`
`118.01°C
`
`70
`
`120
`170
`Temperature (°C)
`
`220
`
`270
`
`Universal V2.3C TA Instruments
`
`-1
`
`-3
`
`-5
`
`-7
`
`Heat Flow (W/g)
`
`-9
`20
`
`Exo Up
`
`Merck Exhibit 2140, Page 9
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`3. Hygroscopicity
`
`
`
`Figure 5. Hygroscopicity of sitagliptin freebase at 25°C.
`
`16.
`
`I assessed the hygroscopicity of the sitagliptin freebase using dynamic
`
`vapor sorption at 25°C. See EX2141 (LNB 60659-112) at 21; EX2146 (Dec. 21,
`
`2001 Preliminary Assessment) at 3–4. As shown in Figure 5, I determined that the
`
`freebase was non-hygroscopic, gaining only 0.125% by weight of water over the
`
`entire range of relative humidities.
`
`4.
`
`Solution Stability
`
`17.
`
`I investigated the stability of the freebase using HPLC to quantify the
`
`freebase’s degradation. In HPLC method that I used, sitagliptin elutes at 9.27
`
`minutes and its major degradation products elute at relative retentions times
`
`(“RRTs”) of 0.17, 0.91 and 1.3. See EX2141 (LNB 60659-112) at 21; EX2146
`
`(Dec. 21, 2001 Preliminary Assessment) at 5–7. The degradation products at
`
`10
`
`Adsorption/Desorption Isotherm
`
`Adsorption
`
`Desorption
`
`0.140
`
`0.120
`
`0.100
`
`0.080
`
`0.060
`
`0.040
`
`0.020
`
`0.000
`
`Weight (% change)
`
`0
`
`20
`
`40
`
`60
`
`80
`
`100
`
`%RH
`
`Merck Exhibit 2140, Page 10
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`RRTs 0.17 and 0.91 are products of amide hydrolysis, while the products at RRT
`
`1.3 are attributable to deamination. These degradative pathways are shown below
`
`and were confirmed by James Qin, a former of colleague of mine at the time from
`
`PR&D—using liquid chromatography mass spectrometry (“LC/MS”). See
`
`EX2149 (June 30, 2004 Monohydrate Report) at 17.
`
`
`
`18. Using a 1.0 mg/mL stock solution of the freebase in HPLC-grade
`
`water, I prepared several sample sets at different pHs to determine the stability of
`
`the freebase in solution. For each of the three stability time points (1, 2, and 4
`
`weeks), I prepared a set of samples at different pHs (2, 4, 6, 8, 10, and water) to be
`
`maintained at 5 different stability conditions: −20°C freezer; 5°C refrigerator,
`
`11
`
`F
`
`F
`
`F
`
`F
`
`F
`
`F
`
`N
`
`N
`
`base
`or
`acid
`
`N
`
`N
`
`CF3
`
`NH2
`
`O
`
`L-224715
`
`
`
`O
`
`N
`
`N
`
`N
`
`N
`
`CF3
`
`F
`
`F
`
`F
`
`F
`
`F
`
`F
`
`elimination (de-amination) degradates
`
`NH2
`
`O
`
`HN
`
`O
`
`N
`
`N
`
`N
`
`CF3
`
`amide bond cleavage
`degradates
`
`O
`
`N
`
`N
`
`N
`
`N
`
`CF3
`
`Merck Exhibit 2140, Page 11
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`25°C/60% RH oven; 40°C/75% RH oven; and 80°C/ambient RH oven. See
`
`EX2141 (LNB 60659-111) at 17.
`
`The solution stability data I obtained are reported in Table 1 and 2, below. Table 1
`reports the relative amount of the freebase calculated as an HPLC area % relative
`to the samples stored at −20°C;
`19. Table 2 reports the absolute area % of the major degradants at RRTs
`
`0.17 and 0.91 (hydrolysis), and 1.3 (deamination). See EX2141 (LNB 60659-117,
`
`-123, -127) at 36, 46, 50; EX2142 (LNB 60659 Supp. Data) at 3–52, 75–113, 114–
`
`52 (HPLC chromatograms); see also EX2146 (Dec. 21, 2002 Preliminary
`
`Assessment) at 5–6 (2-week data); EX2151 (Jan. 2002 MRL Progress Report) at 9
`
`(4-week data).
`
`Table 1. Solution stability (relative area % of sitagliptin freebase).
`
`
`1 wk.
`99.9
`99.3
`114.3
`98.8
`99.1
`98.8
`
`5°C
`2 wks.
`98.2
`99.6
`100.0
`99.9
`100.0
`98.7
`
`
`4 wks.
`99.3
`99.4
`98.9
`100.1
`99.6
`98.2
`
`
`1 wk.
`98.0
`99.4
`97.0
`98.8
`97.1
`96.2
`
`25°C
`2 wks.
`91.7
`100.0
`100.2
`100.3
`95.4
`89.4
`
`
`4 wks.
`88.0
`99.6
`99.1
`100.1
`90.8
`79.4
`
`
`1 wk.
`88.4
`99.5
`114.3
`98.5
`88.2
`83.8
`
`40°C
`2 wks.
`71.5
`100.0
`100.3
`99.1
`77.3
`62.0
`
`
`4 wks.
`54.9
`99.8
`99.7
`97.5
`58.2
`36.7
`
`
`1 wk.
`1.1
`98.4
`94.9
`20.3
`0.5
`2.4
`
`80°C
`2 wks.
`0.4
`97.7
`95.7
`4.4
`0.5
`0.5
`
`
`4 wks.
`0.0
`95.1
`87.9
`0.3
`0.5
`0.5
`
`
`
`Condition
`Water
`pH 2
`pH 4
`pH 6
`pH 8
`pH 10
`
`
`Table 2. Solution stability (relative area % freebase degradants, 80°C samples).
`
`
`Condition
`Water
`pH 2
`pH 4
`pH 6
`pH 8
`pH 10
`
`RRT 0.17
`2 wks.
`8.95
`0.76
`0.51
`7.61
`14.94
`17.83
`
`4 wks.
`4.22
`0.60
`0.83
`8.72
`15.19
`17.91
`
`1 wk.
`6.80
`0.45
`0.34
`6.14
`13.68
`18.43
`
`RRT 0.91
`2 wks.
`19.96
`1.52
`1.00
`20.89
`23.73
`34.25
`
`4 wks.
`22.38
`2.97
`1.93
`22.65
`23.32
`33.63
`
`1 wk.
`18.70
`0.68
`0.43
`17.17
`23.15
`34.98
`
`RRT 1.3
`2 wks.
`68.12
`0
`2.69
`65.21
`58.45
`43.07
`
`4 wks.
`68.30
`0
`5.99
`66.26
`58.73
`43.95
`
`1 wk.
`72.91
`0
`2.33
`44.26
`60.29
`43.80
`
`
`
`12
`
`Merck Exhibit 2140, Page 12
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`
`
`Synthesis & Characterization of the Phosphate, Besylate, and
`Tartrate Salts of Sitagliptin
`
`20. The results of my characterization studies on the sitagliptin freebase
`
`led the DPP-IV project team to conclude that the freebase was not desirable for
`
`further development. In particular, the freebase degraded significantly in solution
`
`at all pHs due to hydrolysis of sitagliptin’s amide bond. The freebase also suffered
`
`from thermally induced deamination at high temperatures in solution and in the
`
`bulk. Accordingly, an early priority of the DPP-IV project team was the
`
`identification of a suitable salt of sitagliptin for further development.
`
`21.
`
`Initial salt formation experiments using sitagliptin were performed by
`
`Vicky Vydra, a named co-inventor of the ’708 patent, who successfully
`
`crystallized phosphate, besylate, and tartrate salts of sitagliptin in December 2001.
`
`I learned of the results of Ms. Vydra’s experiments in or around late December
`
`2001 or early January 2002 from other members of the DPP-IV project team,
`
`including Drs. Michael Palucki and Karl Hansen.
`
`22. The quantities of the sitagliptin salts crystallized by Ms. Vydra were
`
`too small to perform further characterization. Dr. Hansen—another member of the
`
`DPP-IV project team and a named co-inventor of the ’708 patent—undertook
`
`additional salt formation experiments to produce large-scale quantities of various
`
`sitagliptin salts, including the salts that had been initially crystallized by Ms.
`
`Vydra. Over the course of January, February, and March 2002, I—and several
`
`13
`
`Merck Exhibit 2140, Page 13
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`other scientists at Merck including Dina Zhang and Christopher Lindemann—
`
`characterized samples provided by Dr. Hansen in order to provide the DPP-IV
`
`project team with the data necessary to select a sitagliptin salt for further
`
`pharmaceutical development.
`
`23.
`
`In accordance with the customary practice of Merck scientists at the
`
`time, the team kept track of the salts synthesized by Dr. Hansen using references to
`
`his lab notebooks pages. Between January and March 2002, I received and
`
`characterized samples from at least four different batches synthesized by Dr.
`
`Hansen: 70316-025, 70316-031, 70316-035, and 70316-043. See EX2141 (LNB
`
`60659-153, -160, -163, -169) at 91, 109, 111, 137. At this time, the phosphoric
`
`acid salts synthesized by Dr. Hansen were anhydrous.
`
`24.
`
` I reported the results of my characterization studies, as well as the
`
`data gathered by other DPP-IV project team members, in an April 9, 2002
`
`memorandum, a second memorandum dated September 30, 2002, and my February
`
`2002 and March 2002 MRL progress reports, true and correct copies of which may
`
`be found in EX2147, EX2148, EX2152, and EX2153, respectively.
`
`1.
`
`Particle Morphology
`
`25. The particle morphologies of the sitagliptin salts synthesized by Dr.
`
`Hansen were examined using scanning electron microscopy (“SEM”). SEM
`
`images taken by Dina Zhang—a colleague of mine at the time from PR&D who
`
`14
`
`Merck Exhibit 2140, Page 14
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`also worked on the DPP-IV project—are shown in Figure 6, below. See EX2156
`
`(Feb. 28, 2002 Salt Selection for L-224715 and L-221869) at 11; see also EX2141
`
`(LNB 60659-148) at 86 (images taken using optical microscope).
`
`
`
`
`
`
`
`
`
`
`
`
`
`Figure 6. SEM images of the 1:1 DHP (left), besylate (center),
`and tartrate (right) salts of sitagliptin
`
`26. As shown in Figure 6, the 1:1 DHP salt had a flake- or plate-like
`
`morphology. This particle morphology was preferred by the DPP-IV project team
`
`over higher aspect ratio morphologies such as needles from the standpoint of
`
`improved pharmaceutical processability. The superior particle morphology of the
`
`1:1 DHP salt and data on its improved processability were discussed in a February
`
`22, 2002 memorandum sent to me from Dina Zhang, a true and correct copy of
`
`which may be found in EX2159.
`
`27.
`
`In contrast, the besylate and tartrate salts formed particles with less
`
`favorable needle-like or rod-like morphology. The besylate salts had length of
`
`about 10 µm and an aspect ratio of about 5:1. See EX2141 (LNB 60659-148) at
`
`15
`
`Merck Exhibit 2140, Page 15
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`86. The tartrate salts consisted of very long, thin needles, with a length of around
`
`100 µm an aspect ratio of about 50:1. See id.
`
`2. Hygroscopicity
`
`L-224715 H3PO4 Salt
`Adsorption/Desorption Isotherm
`
`20
`
`40
`
`60
`
`80
`
`100
`
`%RH
`
`Adsorption
`
`Desorption
`
`
`
`0.600
`
`0.500
`
`0.400
`
`0.300
`
`0.200
`
`0.100
`
`0.000
`
`0
`
`-0.100
`
`-0.200
`
`Weight (% change)
`
`Figure 7. Absorption/desorption isotherm for 1:1 DHP salt at 25°C.
`
`28. The hygroscopicity of the 1:1 DHP salt of sitagliptin (“L-224715-
`
`006F006”) was assessed using dynamic vapor sorption at 25°C. See EX2147 (Apr.
`
`9, 2002 Pharmaceutical Evaluation) at 3, 6. As shown in Figure 7, the 1:1 DHP
`
`salt is non-hygroscopic, gaining less than 0.5 wt.% water between 5 and 95% RH.
`
`29.
`
`In contrast, the besylate and tartrate salts of sitagliptin were found to
`
`be hygroscopic. The besylate salt converts to a hemihydrate at about 85% RH.
`
`The tartrate salt gains 1.4 wt.% at 15% RH; below this, the salt is an unstable
`
`anhydrate that readily absorbs water under ambient conditions. These findings
`
`were reflected in presentations to the DPP-IV project team, true and correct copies
`
`16
`
`Merck Exhibit 2140, Page 16
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`of which may be found in EX2156 (Feb. 28, 2002 Salt Selection for L-224715 and
`
`L-221869); EX2155 (Feb. 14, 2002 L-224715 Presentation).
`
`3.
`
`Stoichiometry
`
`30.
`
`I observed that the stoichiometry of the phosphate salt synthesized by
`
`Dr. Hansen was likely to be that of a monobasic 1:1 DHP salt based on the fact that
`
`the pH of a 1.0 mg/mL solution of the salt was 5.61. See EX2152 (Feb. 2002 MRL
`
`Progress Report) at 5; see also EX2141 (LNB 60659-153) at 91.
`
`31.
`
`I confirmed the 1:1 stoichiometry of the phosphate salt synthesized by
`
`Dr. Hansen (LNB 70316-043) using HPLC analysis against a standard solution of
`
`the freebase. The average salt factor I measured was 0.804; the theoretical salt
`
`factor for a 1:1 DHP salt is 0.806, thus confirming the 1:1 stoichiometry. See
`
`EX2153 (Mar. 2002 MRL Progress Report) at 1; EX2141 (LNB 60659-169, -178)
`
`at 137, 161. I also determined the ethanol solubility of four different batches of
`
`Dr. Hansen’s phosphate salts (LNB 70316-025, -031, -035, and -043); the values
`
`obtained ranged from 0.181 to 0.196 mg/mL, which confirmed that the batches had
`
`the same polymorphic form and stoichiometry. See EX2153 (Mar. 2002 MRL
`
`Progress Report) at 1; EX2141 (LNB 60659-169, -178) at 137, 161.
`
`4.
`
`Solution Stability
`
`32.
`
`I also investigated the solution stability of samples of the 1:1 DHP
`
`(70316-025), besylate (LNB 70130-347), and tartrate (LNB 70130-359) salts
`
`17
`
`Merck Exhibit 2140, Page 17
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`synthesized by Dr. Hansen. See EX2141 (LNB 60659-142, -143, -153) at 78, 79,
`
`91. Although the goal of the DPP-IV project was to produce a solid oral dosage
`
`form, having a form of sitagliptin with high solution stability was an important
`
`goal to ensure that the API would remain stable during pharmaceutical processing,
`
`as well in situations where the API could become dissolved in solution, such as in
`
`IV formulations or high relative-humidity environments.
`
`Using a 1 mg/mL stock solution of each salt, I prepared a total of 54 samples of
`each salt for 3 stability time points (1, 2, and 4 weeks), 3 different temperature
`conditions (−20°C, 40°C, and 80°C) and 6 different pH conditions (pH 2, 4, 6, 8,
`10, and water); 4 week data for the tartrate salt was not collected due to the
`degradation observed at 2 weeks. Stability was assessed using HPLC and
`calculated as relative area % to the samples stored at −20°C. The solution stability
`for the 1:1 DHP, besylate, and tartrate salts are shown in Table 3–
`33. Table 5, below. See EX2141 (LNB 60659-175, -176, -177) at 158–60
`
`(reporting 4-week data); see also EX2152 (Feb. 2002 MRL Progress Report) at 2–
`
`4.
`
`Table 3. Solution stability of 1:1 DHP sitagliptin salt (LNB 60659-175).
`
`
`Conditions
`Water
`pH 2
`pH 4
`pH 6
`pH 8
`pH 10
`
` Relative Area %, 40°C
`1 wk.
`2 wks.
`4 wks.
`93.9
`89.8
`75.9
`99.9
`98.1
`100.6
`101.8
`93.2
`94.1
`99.0
`101.6
`98.5
`88.2
`81.6
`60.2
`83.1
`62.9
`36.5
`
`Relative Area %, 80°C
`1 wk.
`2 wks.
`4 wks.
`0.0
`0.0
`0.0
`99.0
`100.3
`95.4
`100.9
`93.6
`90.9
`24.9
`5.0
`0.4
`0.0
`0.0
`0.0
`0.0
`0.0
`0.0
`
`18
`
`Merck Exhibit 2140, Page 18
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`Table 4. Solution stability of sitagliptin besylate salt (LNB 60659-176).
`
`
`Conditions
`Water
`pH 2
`pH 4
`pH 6
`pH 8
`pH 10
`
` Relative Area %, 40°C
`1 wk.
`2 wks.
`4 wks.
`85.8
`73.2
`51.9
`99.3
`94.9
`104.5
`101.8
`94.7
`98.3
`102.6
`93.9
`99.5
`85.1
`69.1
`50.0
`74.0
`53.4
`28.9
`
`Relative Area %, 80°C
`1 wk.
`2 wks.
`4 wks.
`0.0
`0.0
`0.0
`97.4
`92.3
`95.9
`97.5
`90.9
`93.2
`23.9
`4.3
`0.0
`0.0
`0.0
`0.0
`0.0
`0.0
`0.0
`
`Table 5. Solution stability of sitagliptin tartrate salt (LNB 60659-177).
`
`
`Conditions
`Water
`pH 2
`pH 4
`pH 6
`pH 8
`pH 10
`
` Relative Area %, 40°C
`1 wk.
`2 wks.
`4 wks.
`89.2
`81.8
`n.d.
`99.4
`99.1
`n.d.
`100.5
`94.5
`n.d.
`99.3
`97.4
`n.d.
`88.0
`78.1
`n.d.
`78.3
`61.6
`n.d.
`
`Relative Area %, 80°C
`1 wk.
`2 wks.
`4 wks.
`0.0
`0.0
`n.d.
`99.7
`97.6
`n.d.
`100.7
`93.0
`n.d.
`26.0
`5.9
`n.d.
`0.0
`0.0
`n.d.
`0.0
`0.0
`n.d.
`
`
`
`
`
`34. The 1:1 DHP, besylate, and tartrate salts of sitagliptin are most stable
`
`between pH 2 and 4; however, the stability at 40°C of the 1:1 DHP salt was
`
`markedly superior to the tartrate and besylate salts. This was attributed to a pH
`
`effect as solutions of the 1:1 DHP salt had lower pH after storage at 40°C.
`
`35. To further demonstrate the superior solution stability of the 1:1 DHP
`
`salt, I measured the degradation of the salts (and the freebase) at a 0.1 mg/mL
`
`concentration in unbuffered water stored at 40°C for four weeks. The results I
`
`obtained are shown in Table 6 below. See EX2141 (LNB 60659-175) at 158. The
`
`19
`
`Merck Exhibit 2140, Page 19
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`data indicated that the 1:1 DHP salt is much more stable in unbuffered water
`
`compared to the freebase and the besylate and tartrate salts.
`
`Table 6. Comparative degradation of sitagliptin salts (60659-175).
`
`
`
`Phosphate
`Freebase
`Besylate
`Tartrate
`
`Rel. Area (%) Hydrolysis
`1 wk.
`2 wks.
`4 wks.
`3.8
`7.8
`19.5
`8.9
`20.4
`47.7
`8.3
`18.2
`41.3
`6.3
`13.7
`n.d.
`
`Rel. Area % (Deamination)
`1 wk.
`2 wks.
`4 wks.
`2.3
`5.1
`9.1
`3.8
`11.8
`28.8
`3.6
`7.4
`25.3
`4.2
`6.1
`n.d.
`
`
`III. SELECTION OF THE 1:1 DHP SALT OF SITAGLIPTIN
`
`36. The DPP-IV project team selected the 1:1 DHP salt of sitagliptin for
`
`further development in late February 2002. This decision was driven the 1:1 DHP
`
`salt’s rare combination of highly favorable properties. First, the 1:1 DHP salt was
`
`more stable than the freebase, as well as the besylate and tartrate salts that were
`
`characterized, especially with respect to stability in aqueous solution. See supra
`
`Table 6. The 1:1 DHP salt was non-hygroscopic and no deliquescence was
`
`observed after bulk storage at 40°C/75% RH for one week. Additionally, the
`
`flake- or plate-like morphology of 1:1 DHP salt was preferable to the needle-like
`
`or rod-like morphology of the besylate and tartrate salts.
`
`37. One of my colleagues, Ivan Santos, remarked that this combination of
`
`favorable properties was “incredible” and “not often” seen in a February 23, 2002
`
`email, a true and correct copy of which may be found in EX2157. I was personally
`
`surprised by the combination of superior properties of the 1:1 DHP salt, as I had
`
`20
`
`Merck Exhibit 2140, Page 20
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`initially expected the tartrate to be selected based on my work on L-221869, a
`
`closely related and structurally similar compound to sitagliptin. See infra Section
`
`IV. I expressed this view to Dr. Hansen in a January 11, 2002 email, a true and
`
`correct copy of which may be found in EX2133. The 1:1 DHP salt’s unique and
`
`rare combination of superior properties made it the clear favorite of the sitagliptin
`
`salts the were considered by the DPP-IV project team and ultimately led to its
`
`selection for further development.
`
`IV. CHARACTERIZATION OF THE L-221869 SALTS
`
`38.
`
`In parallel with sitagliptin, the DPP-IV project team also worked on
`
`closely related lead compound, L-221869. The structure of L-221869 differs from
`
`sitagliptin in that L-221869 has two fluorine substitutions on its left-side phenyl
`
`ring, instead of three. L-221869 and sitagliptin are shown in Figure 8, below.
`
`
`
`
`
`Figure 8. Chemical structures of L-221869 (left) and sitagliptin (right).
`
`39. Following the selection of the 1:1 DHP salt of sitagliptin for further
`
`development, the DPP-IV project also evaluated the suitability of a phosphate salt
`
`of L-221869. In September 2002, using dynamic vapor sorption at 25°C, I
`
`21
`
`Merck Exhibit 2140, Page 21
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`determined that the L-221869 phosphate salt was very hygroscopic, adsorbing 25%
`
`water by weight at 95% RH. The salt gained nearly 5% by weight by 75% RH.
`
`See EX2143 (LNB 26180-132, -135, -137) at 7–9; EX2144 at 5–6; see also
`
`EX2154 (Sept. 2002 MRL Progress Report) at 1–2. The adsorption/desorption
`
`isotherm for the L-221869 phosphate salt is shown in Figure 9.
`
`
`
`Figure 9. Hygroscopicity of L-221869 phosphate salt at 25°C.
`
`40. The hygroscopicity of the L-221869 phosphate salt was in sharp
`
`contrast to the 1:1 DHP salt of sitagliptin, which I had determined was non-
`
`hygroscopic. See supra Section II.B.2. The hygroscopicity of the L-221869
`
`phosphate salt was sufficient to rule this salt out for further development. The same
`
`problem—unacceptable hygroscopicity—had also previously ruled out the
`
`hydrochloride salt of L-221869 for further development. See EX2150 (Dec. 2001
`
`MRL Progress Report) at 1.
`
`22
`
`Adsorption/Desorption Isotherm
`L-000221869 phosphate salt (25 C)
`
`24.988
`
`Adsorption
`
`Desorption
`
`0.364
`
`4.322
`
`0
`
`20
`
`40
`
`60
`
`80
`
`100
`
`%RH
`
`30.000
`
`25.000
`
`20.000
`
`15.000
`
`10.000
`
`5.000
`
`0.000
`
`Weight (% change)
`
`Merck Exhibit 2140, Page 22
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH SHULTZ, PH.D.
`
`V. CHARACTERIZATION OF THE CRYSTALLINE MONOHYDRATE
`OF THE 1:1 DHP SALT OF SITAGLIPTIN
`
`41.
`
`In late March/early April 2003, the DPP-IV project team identified a
`
`new crystal form of the 1:1 DHP salt of sitagliptin: the crystalline monohydrate.
`
`The appearance of this new crystal form—which took place during efforts to scale
`
`up the crystallization of an anhydrous polymorph—was a surprising and
`
`unexpected development to me and the DPP-IV project as a whole, as the team had
`
`spent more than year developing anhydrous forms of the 1:1 DHP salt in aqueous
`
`systems without identifying the monohydrate.
`
`42. The appearance of the monohydrate spurred an intense effort to
`
`characterize the new crystal form and to evaluate its behavior in pharmaceutical
`
`formulations. Numerous scientists from PR&D including myself participated in
`
`this effort. We ultimately determined that the monohydrate unexpectedly exhibited
`
`several advantages over previous formulations using anhydrous forms of the 1:1
`
`DHP salt, in particular, reduced stickiness of both the neat API and in
`
`formulations, as well as improved chemical stability in stress tests using high
`
`temperature, relative humidity, and formaldehyde.
`
`23
`
`Merck Exhibit 2140, Page 23
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`
`
`
`
`DECLARATION OF REBECCA LEIGH